Effortless plugin and play Optimizer to cut model training costs by 50%. New optimizer that is 2x faster than Adam on LLMs.
A unified library of SOTA model optimization techniques like quantization, pruning, distillation, speculative decoding, etc. It compresses deep learning models for downstream deployment frameworks like TensorRT-LLM, TensorRT, vLLM, etc. to optimize inference speed.
🚀 Accelerate inference and training of 🤗 Transformers, Diffusers, TIMM and Sentence Transformers with easy to use hardware optimization tools
DeepSpeed is a deep learning optimization library that makes distributed training and inference easy, efficient, and effective.
Distributed Asynchronous Hyperparameter Optimization in Python
An implementation of model parallel autoregressive transformers on GPUs, based on the Megatron and DeepSpeed libraries
Making large AI models cheaper, faster and more accessible
1 capture since 2026-05-27
